December 13, 2017
CFOs have an opportunity to cut costs, build for the future, and increase data-driven decision making.
I’ve always been surprised at how finance functions treat their own technology spend. Maybe finance executives are more conservative with their own wallets, or maybe they hold themselves to too high a standard. Either way, they tend to invest less in technology than their counterparts in other functions invest—on average, just 13 percent of tech budgets go toward enterprise functions like finance, risk, and compliance, while IT garners 28 percent and sales and marketing gets 24 percent, according to PwC’s 2017 Global Digital IQ Survey.
Some CFOs, however, are rethinking the digital investment in their own function—especially the investment in emerging tech, such as robotic process automation (RPA), artificial intelligence (AI), and blockchain. And it’s not because they merely want to boost productivity and cut costs through automation. They also see ways to create new possibilities for better data-driven decision making beyond the bottom line, thanks to a fresh approach for deploying technology.
From auditing to invoicing and beyond
More on the approach later; first, let’s talk about the possibilities. One that I’m particularly excited about is audit innovation. There’s more to a higher-quality audit than making it more efficient. Think of how external auditors are examining more (if not all) transactions for anomalies throughout the year, changing the audit from a stressful war room busy season event to a lower-key everyday experience. There will be some additional value as more data helps us understand what is happening as it happens, which will provide benefit to the auditor and company management. Ultimately, it is the company’s responsibility to have the process and controls in place, but are they starting to look at evolving their approach as emerging technologies allow for more robust analysis?
Such scenarios will soon be possible with combinations of emerging technologies. Natural language processing has matured to enable AI to ingest a variety of unstructured data including contracts, invoices, and bank statements. Through machine learning, people can begin to train algorithms to detect anomalies or improve data capture from unstructured sources of data, allowing finance professionals to spend more of their time on value-added (and often more personally fulfilling) analyses. And it can all happen in near real time, rather than in big batches after a quarter ends.
For example, PwC has developed and started using a complex algorithm for revenue transactions that helps to match sales orders, invoices, shipping documents, and cash receipts out of a client’s ERP system, and that information helps to identify standard and nonstandard revenue transactions. Our auditors spend more time analyzing the data—by using visualization capabilities—from this match to understand potential risks, to identify areas where nonstandard transactions are being executed and to have more insightful discussions with management about their business. Discussion topics related to nonstandard transactions have included barter transactions and intercompany transactions, as well as potentially higher-risk issues like price overrides and human input error. Finance organizations could conduct the same type of analysis by using their data and some of these emerging technologies.
To be sure, deploying new technologies will not always be as simple as the articles and service providers make it sound. Many times, you will need to establish some basic building blocks, such as understanding what question you are trying to answer, ensuring you have enough data that is organized and labeled, working across the organization with related functions, and developing in-house experience and skills with emerging tech.
A foundation for future dividends
Here’s where the new approach comes into play. If you do it right, those building blocks can reduce costs, as well as prepare your organization for the future. Companies now can use natural language processing, for example, to read contracts to assess compliance with new regulations. This ability helps to build the expertise and the trust in the algorithms, and to organize unstructured data.
Or companies can use RPA to standardize and streamline processes, connecting steps to reduce human interaction in the account reconciliation process, for example. Working together, CFOs and CIOs can build the business case for further investments in machine learning and other emerging technologies.
By adopting these approaches, CFOs can create a win-win situation: cut costs, improve controls, mitigate risk, build for the future, and increase data-driven decision making. That’s the kind of opportunity that should pass any financial analysis and allow finance functions to use their technology spend more strategically.